SciTransfer
EITHOS · Project

AI-Powered Detection System for Deepfakes and Online Identity Theft

digitalTestedTRL 5

Imagine a digital security guard that can tell if a video call or a voice message is a fake created by AI. It also scans the dark web to find where stolen identities are being traded. The goal is to give police better tools to catch scammers and teach regular people how to protect their data.

By the numbers
12
consortium partners
5
countries involved
17
total deliverables
The business problem

What needed solving

Criminals are using sophisticated AI tools like deepfakes and botnets to steal identities, leaving police and citizens unable to distinguish real media from fake.

The solution

What was built

A software toolset for audio/video deepfake detection and botnet detection, alongside a public-facing observatory for citizen education.

Audience

Who needs this

Law Enforcement AgenciesCybersecurity firmsDigital identity verification companiesGovernmental policy makers
Business applications

Who can put this to work

Cybersecurity
mid-size
Target: Anti-fraud software provider

If you are a software provider dealing with the rise of AI-generated scams — this project developed a toolset for audio and video deepfake detection that improves the accuracy of fraud investigations.

Banking
enterprise
Target: Digital banking security department

If you are a bank dealing with sophisticated identity theft in e-banking — this project developed a system for detecting fake botnets and structured knowledge extraction that helps identify fraudulent patterns.

Legal Services
SME
Target: Digital forensics firm

If you are a forensics firm dealing with the challenge of presenting e-evidence in court — this project developed research on the legal framework for AI remote technologies that supports the admissibility of digital evidence.

Frequently asked

Quick answers

What is the cost or pricing for the EITHOS toolset?

Based on available project data, no pricing or cost information is provided as this is a research project funded by the EU.

Can this system be scaled to an industrial level?

The project involves 12 partners and is developing a robust software toolset, but based on available project data, the current scale is for Law Enforcement Agencies and citizen education.

Who owns the IP and what are the licensing terms?

Based on available project data, specific licensing terms are not mentioned, though the project involves a consortium of 12 partners including SMEs and universities.

How does the system integrate with existing police tools?

The project includes integration planning for the toolset into the EITHOS Observatory specifically designed for Law Enforcement Agencies.

What is the timeline for the final rollout?

The project period is from 2022-10-01 to 2025-09-30.

Consortium

Who built it

The consortium is diverse, consisting of 12 partners across 5 countries. It has a balanced mix of 3 universities, 3 research organizations, and 4 other entities, with a 17% industry ratio including 4 SMEs. This suggests a strong academic foundation with a targeted but limited transition to commercial application.

How to reach the team

Contact ETHNIKO KENTRO EREVNAS KAI TECHNOLOGIKIS ANAPTYXIS in Greece

Next steps

Talk to the team behind this work.

Contact us to explore licensing opportunities for deepfake detection tools.